Look at today’s polls and you’d believe that “Prime Minister Farage” is inevitable. Labour’s vote share has crashed since their election victory; His Majesty’s Opposition, the Tories, are in freefall and Farage’s limited company, Reform UK, is surging ahead of both parties.
But what if I told you that the biggest force in UK politics wasn’t Labour voters, Tory voters or even Reform voters…but the hidden power of undecided voters?
Eager Reformers, disaffected Tories and angry Labour critics shape today’s headlines, but there are possibly millions of voters who do not yet know which party they could vote for, and they have the power to shape the next election. Are their stories being told in polling today?
How opinion polling works
Polling companies (such as Ipsos, YouGov and More In Common) periodically ask a sample of the population some questions relating to voting intention: “how likely are you to vote?”, “who would you vote for if an election was held tomorrow?”, etc…
This sounds straightforward. But the headline number that you see published at the end doesn’t reflect the full picture (and for good reason).
The samples that the polling companies choose are intended to be broadly representative of the eligible voting population (in terms of age, geography, ethnicity, class, etc…). Getting good, regular and unbiased samples is slow, difficult and expensive. So to account for this, pollsters will weight different demographics within the sample to offset these biases and “synthesise” a more representative picture, and turn this into a headline voting intention.
This picture acts as a snapshot of how the popular vote might look if a general election were held today. By calculating voting intention periodically, we are able to track how the country “feels” about the existing political parties over time, and whether political party messages are cutting through with target voters.
Different pollsters have different methodologies for arriving at that headline figure, but what is common to most of them is how they treat undecided voters within the samples. Ever notice how these headline numbers for all the parties tend to add up to 100%, yet if you asked around you’d definitely find people who were unsure who to vote for? That’s because undecideds are normally excluded from the figure. There is a reasonable justification for this though: accurately modelling for undecideds is especially tricky without polling that targets them explicitly (which is again expensive and slow): this is needed because we’d need to get a picture of how these voters have acted at past elections (if they’ve voted at all\!) to get any reading on how they might vote in a future election.
None of this is to say that we should ignore opinion polling, and we especially shouldn’t ignore unfavourable polling just because of possible inadequacies in modelling. What we need to remember however \-- particularly in the context of making predictions about the future based on this polling \-- is that excluding undecided voters leads to polling results that disproportionately amplify the voices of those who are already most politically engaged at that point in the parliamentary cycle: this tends to be voters that are angry with the current government (see polling in 2010-2015 during the Cameron ministry and 1979-1983 in Thatcher’s first ministry for polling that illustrated governments in peril who would go on to win stronger majorities at the subsequent election).
A special voting demographic?
Something often unreported is a peculiar characteristic of elections since the Thatcher era. All have had one crucial thing in common: the leading party always wins amongst the 45-54 age group. This is one of the most consistent voting patterns in modern UK political history\!
Thatcher? Won the 45-54s. Blair? Won the 45-54s. Cameron? Won the 45-54s. Starmer? You guessed it: won the 45-54s.
What makes this demographic interesting, though? If it’s a simple coincidence then it’s just a piece of trivia you could whip out to impress politics nerds. But there are qualities to this group of people that are worth recognising.
These voters are middle aged and so are likely to have met (or are close to meeting) one or more common life goals, such as:
Getting on the property ladder (giving them access to capital, and providing a stake in policies that help maintain homeownership) Raising children (hence having a stake in childcare and schooling policies) Amassing savings through investments and considering future retirement (providing a stake in pension policies)
As a result of this \-- and also being young enough to still be politically fluid, not having locked in lifelong voting habits \-- this demographic is the most fertile ground for swing voters, as they are the most likely to be swayed by broad, favourable policy platforms (rather than narrow, single-issue campaigns).
Here are the percentages from Ipsos for the Conservatives, Labour and Lib Dems going back to 1979:
| | 2024 | 2019 | 2017 | 2015 | 2010 | 2005 | 2001 | 1997 | 1992 | 1987 | 1983\\ | 1979\\ | | :---- | :---- | :---- | :---- | :---- | :---- | :---- | :---- | :---- | :---- | :---- | :---- | :---- | | Con | 22 | 46\ | 43\ | 36\ | 34\ | 31 | 32 | 31 | 47\ | 45\ | 44\ | 46\ | | Lab | 36\ | 28 | 40 | 33 | 28 | 35\ | 41\ | 41\ | 31 | 31 | 27 | 35 | | Lib | 14 | 14 | 7 | 8 | 26 | 25 | 20 | 20 | 18 | 23 | 27 | 15 |
(\ indicates the election winner) (\\ these years used a slightly larger age bracket of 35-54, but the pattern still holds)
After consistently being the kingmakers of elections, where have these voters gone today? They’re sitting squarely on the fence, not prepared to back any party.
A recent poll by Find Out Now gave a headline intention of 25% for Labour, 18% for the Conservatives and 29% for Reform UK. These are pretty dire figures for both Labour and the Conservatives. But when looking at the unadjusted voting intention numbers, we can see that *undecideds make up a plurality across all voters under 75 years old. And in fact, the 40-54 age group (slightly wider than the group I showed above, but the closest approximation) is the group with the highest percentage of undecided voters, at 43%\!
Consequently, even small shifts in this group’s voting intention can produce wildly different headline results from week to week.
This also represents an underlooked aspect of current political analysis: why are so many people unsure of who to vote for?
Look around you today and you’ll probably agree: there is an overwhelming sense of apathy from voters of all stripes. They have been exhausted by one of the most politically tumultuous and socially fractious periods in our recent history: from the financial crisis to austerity; from the EU referendum to the chaotic Brexit negotiations; from Covid lockdowns and partygate through to the Conservative Party psychodrama of the last parliament, the public are distrustful of promises made by politicians of any rosette. Those hopeful enough to still see improvement in their lives are waiting to feel real, material change before they are willing to place their trust in anybody.
Right now the polls reflect the loudest, most motivated and politically active voices. With this much apathy and discontent however, there cannot be a clear picture of how the next election is shaping up to be unless we understand where the undecideds are going. Are they quietly willing to give the current parties a chance, or are they expecting to sit the next one out? That choice can sway the result dramatically. The political battlefield is being fought over a “once bitten, twice shy” electorate, so commentators and political parties alike should take care in understanding the real state of play.
What this means for the major parties
Labour, the Conservatives, and Reform may need to take seriously what is being loudly voiced in the pub, but they equally must consider what is being quietly discussed and believed in households. It would be unwise to assume that voters will “return to the mean” and simply pick either Labour or the Conservatives, but this high degree of uncertainty in the electorate offers opportunity to all players.
For Labour, there is a possibility that undecided voters are patiently waiting to see what they achieve, and that they will be judged on the material successes or failures they deliver. Given that voters don’t like to associate with “losers” or negative press, it perhaps makes sense that even those who voted Labour in 2024 are not prepared to say with certainty that they’d do so again: they don’t want egg on their faces for their choices, and certainly don’t want to defend a seeming lame duck. Perhaps we might consider this a first instance of a “shy Labour” effect?
In this case, pivoting to take on Reform directly comes with notable risks: could adopting hardline immigration rhetoric end up appalling undecideds whilst failing to peel off Reform’s support? Of course they shouldn’t allow Farage to dominate the narrative, but they should take great care not to do so in ways that spurn voters who see his rhetoric as part of the exhausting politics of the last decade. He’s succeeded in turning key issues that they care about \-- public services, wages and trust in institutions \-- into wedge issues by blaming immigration. But Labour doesn’t need to fight him on these terms. The solution is not to simply be “tougher” than the Tories or Reform on immigration: it is to render the immigration debate irrelevant by visibly improving living standards with common-sense policies. His culture war isn’t to be won, but to be diffused altogether.
For the Tories, the old concept of the “shy Tory” was decimated in 2024, as these purported voters simply didn’t turn up. Badenoch’s current strategy has been to try and out-Farage Farage, to dismal results. Where do traditional One Nation Conservatives (a venerable and solid voter block within Conservative history) go, if the party they’ve historically voted for seems to show no interest courting the political centre any more? I’d wager that many of these undecideds are in fact ex-Tory voters who were waiting for the parliamentary psychodrama to end and hoped the party would soon regain sanity in opposition. That doesn’t seem to be the case for the current leadership. If undecided voters are unimpressed by the current Tory Party, they too risk losing undecided voters to apathy whilst losing their own motivated base to Reform.
As for Reform itself: it has done well to secure the amount of support that it has in such a short space of time into this parliament. This support-base seems engaged and hungry for its success, which is a far cry from what Labour and the Tories are experiencing. They are in poll-position to be significant agitators to the biggest political players.
But if undecided voters are repulsed by the politics that Farage is selling, his populist single-issue brand could unintentionally be a limiting factor. A serious political party requires a broader policy platform than just immigration, and this will inevitably open him up to greater scrutiny. If those 45-54 year old swing voters are distrustful of him AND see him as a credible threat in the next election, this could incentivise tactical voting. As we’ve seen in 2024 with the Tories: if the electorate wants you out, they are prepared to tactically vote to do so. Hence Farage’s biggest concerns will be the impact of his rhetoric outside of his own echo-chambers, and the extent to which his less popular policies are scrutinised in public (in particular with respect to healthcare and the economy).
What Comes Next?
All parties hence have a vested interest in interrogating why the electorate is disengaged and disinterested. Voters need to be given a reason to care about politics: a reason to participate in our democracy. So what can be done?
Political education and awareness
There is a distinct lack of civics education in the UK, which means that people’s understanding of politics is often based on gaps in knowledge of how political processes work. What is your MP meant to be doing? What do your councillors do, and why should you care about local elections?
Parties should be doing a better job of justifying their existence by actually providing that missing civics education for the electorate: explain exactly what it is that politicians do, how it helps us and how to be involved with the process. This also requires amplifying the importance of local politics and getting the public to truly understand what their councillors are doing for them: after all, local council decisions are often the most immediate political decisions that people will experience day-to-day, even if they are politically disengaged.
Make policies personal
Rhetoric dictates how parties position themselves to each other, yet policy is the ultimate distinction between any two parties. But policy is often so abstract that people don’t see how it is meant to affect their daily lives. How will “Policy X” help me afford my groceries? How will “Policy Y” help me raise my child? How will “Policy Z” help me get a home? Parties must spend time not only developing their policies, but pitching and framing them for the electorate in clear, actionable ways that are easy to relate to their own personal circumstances.
Rebuilding trust
The trust deficit in politics is immense at the moment, meaning work needs to be done to tackle disillusionment. For policy discussions to have any salience, politicians need to be trusted on what they say and what they do. To be trusted, they need to be seen speaking the truth in public; to acknowledge with humility when they make mistakes and to actively redress past mistakes. This means showing results to the electorate. In order to believe that parties are different, we need not just words, but actions to demonstrate that they are serious. For the government, this means delivering on their manifesto promises and listening to public concerns when they arise. For the opposition this means re-engaging with experts to shape a credible policy platform and building a front bench that can effectively hold the government to account on its faults.
Conclusions
Opinion polls tell a story, but an incomplete* one. Undecideds are a key, underlooked demographic for everyone (pollsters, political commentators and political parties alike) to take notice of, and these voters’ apathy is something that actively requires investigating and counteracting.
Don’t fall for the belief that the next election is already decided \- early polling will snapshot current moods and loud discontent, but it deliberately suppresses the voices of those reserving judgement on what our parties are going to do to win them over. In that time, we must look at how to amplify constructive engagement with politics for the disillusioned (and not uncritically amplify the voices of the already engaged and entrenched).
The battle for the soul of the country is that of re-engaging a shy, distrustful electorate with serious politics aimed at making their lives better.
Our politicians have four years to rise to this challenge.
The question is: will we hold them to it?